Abstract The long-term goal of this project is to study mechanisms of pituitary control of ovarian and bone function in aging women. Normal ovarian function is dependent on follicle-stimulating hormone (FSH), a pituitary derived heterodimeric glycoprotein consisting of a ?-and a ?-subunit. Both the subunits are glycosylated with two N- linked sugar chains in each subunit. This fully glycosylated form is designated FSH24. Glycosylation plays a major role in secretion, serum half-life and biological actions of FSH. It is known that glycosylation of pituitary gonadotropins is also estrous/menstrual cycle- and age-specific. Biochemical and physiological studies in several species have identified unique hypo-glycosylated variants consisting of sugar chains only in the ? but either one or none on the ? subunit. These are known as hypo-glycosylated FSH glycoforms and designated as FSH21, FSH18 or FSH15. Most importantly, the ratio of hypo- to fully-glycosylated FSH forms is found age- dependent, with high levels of fully-glycosylated variant predominantly present in peri/post-menopausal women and may contribute to the aging-associated bone loss. However, the distinct in vivo biological functions of these FSH glycoform variants are unknown in normal and aging ovarian and bone physiology. The central hypothesis is that glycosylation on FSH is an age-related switch that changes target tissue specificity from ovary to bone. This hypothesis will be tested using genetically engineered mouse models in two specific Aims. In Aim 1, we will test ovarian development and function progressively with aging using Fshb null mice expressing individual glycosylated forms of FSH. This genetic strategy will allow us to test systematically the in vivo biological actions of each glycosylated hFSH variant in ovarian physiology in the absence of endogenous mouse FSH. In Aim 2, first, we will use the FSH glycoform-expressing mice and test bone development as a function of aging. To unequivocally test the direct FSH actions on bone, in one approach, we will engineer mice in which Fshr will be selectively deleted in osteoclasts by a Cre-lox approach. We will develop a second line that permits creating temporal loss of FSH at desired times. Functional analyses with these genetically altered mouse models will identify distinct biological actions of FSH variants in vivo during ovarian aging and by extrapolation in human ovarian senescence. These novel mouse models will also allow us to directly test whether aging-associated bone loss is dependent on FSH ligand or FSH receptor-mediated signaling in osteoclasts in the bone. Our studies may unravel a novel phenomenon of age-dependent N-glycosylation switch on a pituitary glycoprotein hormone that results in alterations in target tissue specificity (ovary versus bone) and may potentially lead to new therapeutic options for intervention of bone loss in aging women.